The ultrasonic cleaning method was applied to elute the particulate matter (PM) retained on leaf surfaces after PM was eluted by the conventional cleaning methods (water cleaning only or water cleaning plus brush cleaning). The methodology can help to improve the estimation accuracy for PM retention capacity of leaves.
Based on the conventional cleaning methods (water cleaning (WC) + brush cleaning (BC)), this study evaluated the influence of ultrasonic cleaning (UC) on collecting various sized particulate matter (PM) retained on leaf surfaces. We further characterized the retention efficiency of leaves to various sized PM, which will help to assess the abilities of urban trees to remove PM from ambient air quantitatively.
Taking three broadleaf tree species (Ginkgo biloba, Sophora japonica, and Salix babylonica) and two needleleaf tree species (Pinus tabuliformis and Sabina chinensis) as the research objects, leaf samples were collected 4 days (short PM retention period) and 14 days (long PM retention period) after the latest rainfall. PM retained on the leaf surfaces was collected by means of WC, BC, and UC in sequence. Then, retention efficiencies of leaves (AEleaf) to three types of the various sized PM, including easily removable PM (ERP), difficult-to-remove PM (DRP), and totally removable PM (TRP), were calculated. Only around 23%-45% of the total PM retained on leaves could be cleaned off and collected by WC. When the leaves were cleaned through WC+BC, the underestimation of the PM retention capacity of different tree species was in the range of 29%-46% for various sized PM. Almost all PM retained on leaves could be removed if UC was supplemented to WC+BC.
In conclusion, if the UC was complemented after the conventional cleaning methods, more PM on leaf surfaces could be eluted and collected. The procedure developed in this study can be used for assessing the PM removal abilities of different tree species.
The abilities of different tree species to remove PM from ambient air can be assessed through quantifying the mass of PM retained on leaf surfaces. To achieve this objective, the subtraction method1,2, the membrane filter method3,4,5, and the elution-weighing method coupled with particle size analysis6 have been applied to quantitatively estimate the mass of PM2.5 (diameter ≤ 2.5 µm), PM10 (diameter ≤ 10 µm) or total suspended particulate (TSP) retained on leaves. However, the accuracy of these methods basically depends on their performance in collecting PM retained on the leaf surfaces. At present, the conventional leaf cleaning method used in related studies often includes one or two steps, namely only water washing (soak and rinse leaves using deionized water)3,7 or plus brushing5,8,9. However, some studies10,11 have demonstrated that PM on leaf surfaces could not be completely eluted by the conventional cleaning method. As ultrasonic cleaning has the advantages of high speed, high quality, and little damage to the surface of the object, it has great potential to be used to collect the PM retained on leaf surfaces with complex microstructures. At present, ultrasonic cleaning has been applied in some studies to collect PM retained on leaf surfaces (i.e., put the leaves into deionized water, and use the ultrasonic cleaner to elute PM)12,13. However, this method is only used as a supplement to a leaf cleaning method, while it is not known whether the ultrasonic cleaning has a positive effect on collecting PM from leaf surfaces and its optimum operating parameters are also not clear. Our previous research has shown that the PM retained on Ginkgo biloba leaf surface could be completely eluted without destroying the leaf surfaces, if a proper ultrasonic cleaning procedure was supplemented to the conventional cleaning method11. However, the stability and general applicability of the ultrasonic cleaning parameters (ultrasonic power, time, and other information) to different plant species experiencing different dust retention periods are still not clear.
Currently, the mass of PM2.5, PM10, or TSP on unit leaf area has often been utilized to evaluate the abilities of different tree species to remove PM from ambient air14,15. Under the natural condition, the PM retained on leaf surfaces can be classified into two parts: the first part is the PM that can fall off leaves due to the effects of wind and rainfall, while the other part is the PM that is tightly adhered to leaf surfaces and cannot be easily washed off by rainfall. However, few studies have focused on the mass of both types of PM on leaf surfaces. In addition, the PM retention periods of leaves in different studies differ enormously. Thus, the comparability of the results of these studies will be poor, if the mass of PM retained on unit leaf area is adopted to assess the PM removal abilities of trees16. Consequently, the PM retention efficiency (the mass of PM retained on unit leaf area per unit time), as an alternative, was proposed to evaluate the PM purification effects of urban trees5,17. In general, there is still a lack of research in this aspect. It is extremely necessary to carry out relevant studies for different tree species to provide methodological basic and data support for assessing the PM removal abilities of different tree species accurately.
Here, three broadleaf tree species (G. biloba, Sophora japonica, and Salix babylonica) and two needleleaf tree species (Pinus tabuliformis and Sabina chinensis) were selected to evaluate their PM removal abilities under two PM retention periods. The leaf sampling site was in Xitucheng Park (39.97° N, 116.36° E), located in an area with heavy pollution in Beijing. The three specific objectives of this study were: (1) to assess the efficiency of different leaf cleaning methods (water cleaning (WC), brush cleaning (BC), and ultrasonic cleaning (UC)) in eluting the PM on leaves, (2) to verify the effect of ultrasonic cleaning on eluting PM, and (3) to assess the retention efficiency of different tree species to PM1, PM2.5, PM5, PM10, and TSP.
1. Leaf Collection, Elution and Mass Measurement of PM
2. Measurement of PM Size Distribution and Leaf Area
3. Data Presentation and Analysis
The PM retained on leaf surfaces had two types under natural conditions. The PM falls off easily by rainfall and wind under natural conditions is defined as the easily removable particulate matter (ERP). This type of PM was represented by the PM eluted by WC in this study. The PM that tightly adheres to leaf surfaces and cannot be easily washed off by BC and UC is defined as the difficult-to-remove particulate matter (DRP). This kind of PM cannot be eluted by natural rainfall and wind.
There was a significant difference among the five tree species in the mass proportions of various-sized PM eluted by different cleaning steps. The result showed that a large number of different sized PM were eluted from leaf surface by WC (Figure 1, Figure 2, Figure 3, Figure 4, and Figure 5). The average eluted proportions (ERP) of various-sized PM of the five tree species were 31% and 35% under the SDR and LDR, respectively (Figure 6).
In addition, the WC showed a stronger effect on eluting PM on leaves of needleleaf tree species (P. tabuliformis and S. chinensis), especially for S. chinensis under the LDR period. As such, the elution percentage of WC was significantly higher than that of BC and UC (P < 0.05) for all sized PM except PM>10. After cleaning the leaves by BC, large fractions of various sized PM were also eluted, which were 28% and 29% under the SDR and LDR periods, respectively. Similar to WC, the most distinct elution effect of BC was observed for S. japonica. The elution percentage of BC was significantly higher than that of WC (P < 0.05) for all fractions of PM under both the SDR and LDR periods. Moreover, the elution effect of BC was significantly higher than that of UC (P < 0.05) except PM with diameter < 5 µm (Figure 6). Although a large fraction of PM could be eluted from leaf surface by WC + BC, some PM with smaller diameters still adhered on leaf surfaces. Subsequently, when UC was applied to clean leaves, the residual PM retained on leaf surfaces were eluted completely (Figure 1, Figure 2, Figure 3, Figure 4, and Figure 5), and the elution proportions were 41% under SDR and 36% under LDR (Figure 6). In addition, the elution percentage of small sized PM was higher when the UC was applied. Consequently, the mass of PM would be underestimated apparently, if the conventional elution method was only adopted to elute the PM on leaves. Especially for S. babylonica, the average eluted proportions of PM in all diameter classes would be underestimated by 46%, which was higher than that for P. tabuliformis (43%), G. biloba (42%), S. japonica (31%), and S. chinensis (29%).
The AEleaf of different types of PM of the five tree species is shown in Table 1. There was a great difference in the retention efficiency calculated by two different methods. Compared with the result estimated by equation (2), the AEleaf calculated by equation (3): retention efficiency (mg/m2·d-1) = the mass of the PM on a unit of leaf area (mg/m2) / the dust retention duration (d) was about 5 times higher. Especially for S. japonica, the PM1 of ERP calculated by equation (3) was 18.94 times higher than that calculated by equation (2). In this study, for ERP, the TSP AEleaf of different tree species varied between 12.69 and 34.69 mg·m-2·d-1 and decreased in the following order: P. tabuliformis > S. babylonica > G. biloba > S. japonica > S. chinensis. While in a previous study, the TSP AEleaf of different tree species varied between 35.27 and 85.79 mg·m-2·d-1 and decreased in the following order: S. japonica > S. chinensis > S. babylonica > P. tabuliformis > G. biloba. The retention efficiency of different tree species in retaining various sized PM of different types (ERP, DRP, TRP) could also vary. In this study, S. japonica exhibited the highest AEleaf in retaining the TRP of PM1 and PM2.5, which were 4.3 and 21.91 mg/m2·d-1, respectively. S. babylonica had the highest AEleaf in retaining the TRP of PM5 (40.98 mg/m2·d-1) and PM10 (62.01 mg/m2·d-1). In addition, S. chinensis could retain more ERP of PM1, PM2.5, and PM5 than the other tree species.
Figure 1: The particulate matter residue on leaves of Ginkgo biloba after different elution steps. A and B stand for the upper and lower sides of leaves of Ginkgo biloba. Different numbers stand for different elution steps. (1: without cleaning; 2: single water cleaning; 3: water cleaning + brush cleaning; 4. water cleaning + brush cleaning + ultrasonic cleaning). Please click here to view a larger version of this figure.
Figure 2: The particulate matter residue on leaves of Pinus tabuliformis after different elution steps. A and B stand for the concave and convex sides of leaves of Pinus tabuliformis. Different numbers stand for different elution steps. (1: without cleaning; 2: single water cleaning; 3: water cleaning + brush cleaning; 4. water cleaning + brush cleaning + ultrasonic cleaning). Please click here to view a larger version of this figure.
Figure 3: The particulate matter residue on leaves of Sophora japonica after different elution steps. A and B stand for the upper and lower sides of leaves of Sophora japonica. Different numbers stand for different elution steps. (1: without cleaning; 2: single water cleaning; 3: water cleaning + brush cleaning; 4. water cleaning + brush cleaning + ultrasonic cleaning). Please click here to view a larger version of this figure.
Figure 4: The particulate matter residue on leaves of Salix babylonica after different elution steps. A and B stand for the upper and lower sides of leaves of Salix babylonica. Different numbers stand for different elution steps. (1: without cleaning; 2: single water cleaning; 3: water cleaning + brush cleaning; 4. water cleaning + brush cleaning + ultrasonic cleaning). Please click here to view a larger version of this figure.
Figure 5: The particulate matter residue on leaves of Sabina chinensis after different elution steps. A and B stand for the conical and scale-form leaves of Sabina chinensis. Different numbers stand for different elution steps. (1: without cleaning; 2: single water cleaning; 3: water cleaning + brush cleaning; 4. water cleaning + brush cleaning + ultrasonic cleaning). Please click here to view a larger version of this figure.
Figure 6: The mass proportions of the various sized particulate matter retained on the leaves of different tree species. A and B stand for short (SDR) and long (LDR) dust retention periods, respectively. WC, BC, and UC stand for the single water cleaning, brush cleaning, and ultrasonic cleaning, respectively. Data are mean ± SE. Different letters (a, b, c) above the data bar indicate significant difference (P < 0.05) among the different cleaning steps on the elution percentages of the different-sized particles under different dust retention periods, according to the Duncan test. Please click here to view a larger version of this figure.
Table 1: The mass of different types of PM retained on unit leaf area. ERP, DRP, and TR stand for the easily removable dust retention capacity, the difficult-to-remove dust retention capacity, and the total dust retention capacity, respectively. Equation (2): retention efficiency (mg/m2·d-1) = the subtract mass of the PM on unit leaf area under the LDR and SDR periods (mg/m2) / the dust retention duration between the LDR and SDR periods (d); Equation (3): retention efficiency (mg/m2·d-1) = the mass of the PM on a unit of leaf area (mg/m2) / the dust retention duration (d).
Accurate and proper collection of the PM retained on leaf surfaces is the basis for assessing the PM removal abilities of different tree species. However, the conventional cleaning method (WC or plus BC) cannot completely remove the dust on leaf surfaces, which has been confirmed by scanning electron microscopy10. This was further demonstrated clearly by the present study (Figure 1, Figure 2, Figure 3, Figure 4, and Figure 5). Our study shows that, if only WC was applied to clean leaves, the PM on leaf surface would be underestimated by about 69% and 65% under the SDR and LDR periods, respectively. That is to say, a short period of rainfall with a certain intensity could only elute 31% and 35% of PM from leaf surfaces. In addition, previous studies have shown that a short and heavy rainfall could only elute 50% and 62% of the PM from the leaves of Ligustrum lucidum and Viburnum odoratissimum, respectively18. However, influence of rainfall on eluting the PM retained on leaf surfaces of P. tabuliformis was not evident. Consequently, under natural conditions, only a small part of PM on leaf surface could be eluted by rainfall. When both WC and BC were applied to clean leaves, the elution percentage of PM would also be underestimated by about 41% and 36% under the SDR and LDR periods, respectively. However, more PM could be eluted and collected from leaf surface after supplementing UC to WC+BC for cleaning leaves. Thus, for unbiased and accurate quantification of PM retained by leaves, it is necessary and crucial to add UC to the conventional leaf cleaning method.
At present, most studies using the retention capacity of PM retained on leaf surfaces to assess the particulate matter removal abilities of trees. Although this indicator is convenient for assessing the PM removal abilities under the same retention duration, there will be great difference in the retention capacities of the same tree species under different dust retention durations. Thus, it was proposed in some studies that the retention efficiency (the mass of PM retained on a unit of leaf area per unit time) should be applied to assess the plant PM removal abilities, as this can eliminate the evaluation deviation of PM removal ability due to the difference in dust retention duration. However, these studies neglected the fact that only a small part of PM on leaf surface can be eluted by rainfall. In addition, this method could result in five times overestimation of the PM removal abilities of trees, according to the results of this study (Table 1). For this reason, the calculation method of equation (2) should be applied to accurately assess the PM removal abilities of trees.
When PM elution method proposed in this study was used to determine the mass of PM on leaf surfaces, each experimental step must be accurate and avoid the errors caused by human factors as much as possible. For example, the number of leaves for each experiment depends on the specific situation, and it should be determined by the specification of experiment instrument, the dust retention duration, ultrasonic parameter, and other factors. Taking ultrasonic parameter as example, the ultrasonic cleaning duration and the centrifuge speed should be determined by multiple preliminary experiments to ensure the experimental error is within the acceptable range. In addition, the amount of deionized water used for elution should also be adjusted according to the mass of PM. Besides, each small beaker filled with eluent, while it is drying in the oven, should be covered with a piece of clean filter paper to prevent dust pollution. In brief, each step in the experiment must be carefully operated in order to repeat the proposed method accurately.
It is extremely necessary and crucial to supplement the ultrasonic cleaning procedure to the conventional leaf cleaning method, so that the PM removal abilities of trees can be assessed more accurately and quantitatively. In order to accurately compare the PM removal abilities of different tree species experiencing different dust retention durations, the retention efficiency should be calculated using the method (equation (2)) proposed in this study. Our proposed comprehensive protocol will be useful for evaluating the PM purification abilities of urban trees and forests in a precise, unbiased, and accurate manner.
The authors have nothing to disclose.
This work was supported by the Fundamental Research Funds for the Central Universities (2017ZY21) and the National Natural Science Foundation of China (21607038).
MSA2258-1CE-DU ten-thousandth scale | Sartorius Scientific Instruments (Beijing) Co., Ltd. | MSA2258-1CE-DU | precision: 0.01 mg |
The IS13320 laser granularity instrument | Beckman Coulter, Brea, USA | IS13320 | working conditions: liquid/power samples; particle size range of measurement: 0.017-2000 μm |
Epson Twain Pro high-quality scanner | Seiko Epson, Nagano, Japan | expression1680 | |
Automatic image analysis software WinRHIZO | Regent Instruments Inc., Quebec, Canada | WinRHIZO Pro 2013a |